Subjects
After gaining approval from local research ethics com
mittees, we recruited patients from three general prac
tices in east London known to have a high proportion
of Bangladeshi patients. Using computerised diabetes
registers where available, and otherwise by manual
search of case notes, we identified patients with
diabetes and approached them to request a tape
recorded interview. Recruitment was usually by letter followed up by telephone call, but in one practice we
recruited opportunistically through practice reception
ists when patients came to book appointments or col
lect prescriptions. In all, 40 of the 44 Bangladeshi
subjects we approached agreed to be interviewed.